InstaMem (Instant Memory) is a simple, fast, and personal memory assistant. Just type a word, or even part of a word, and instantly surface memories connected to people, places, events, or ideas — all organized by tags you define.

1. INSTANT Search — Zero-latency results and Offline use, uses local data storage sync'd with Supabase DB when online [implemented 0.2.0]
2. Customizable Memories — Users define content and custom tag categories [implemented 0.2.0]
3. AI Assisted Memory Management — Users can use natural language to update or create memories [planned 0.5.0]
4. Advanced Tag Handling — Smart tag management and search refinement [planned 0.3.0]

Learn more in Usage Doc

• Framework: React 19, Next.js 15 with App Router
• Languages: Typescript, shell/bash
• UI: ShadCN UI + Tailwind CSS
• Database: Supabase (PostgreSQL) on the cloud, sync'd locally for Offline use.
• Offline Support: PWA + IndexedDB + localforage + Fuse.js search
• Authentication: Supabase Auth, RLS, Github + Google Login
• Deployment: Vercel

WIP - see Roadmap

npm install
npm run db setup
npm run dev

For complete setup instructions, development commands, and workflow details, see Development Guide.

See Architecture Documentation for detailed system design

This repository contains the React frontend. The backend (InstaMem server) will be a separate Python repository using LangChain for AI-powered memory processing.

📚 Complete Documentation Index - Comprehensive guide to all project documentation

Quick Links:

• Feature Roadmap - What's built and what's planned
• Development Guide - Complete setup and workflow
• System Architecture - Technical design overview